PLMMSE Channel Estimation for MIMO Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current channel estimation methods in multi-carrier MIMO systems, such as LS, DFT-Based, and LMMSE, face challenges in accurately estimating channel gains due to noise and interference between antennas, particularly when prior statistic information is hard to obtain, leading to suboptimal performance.
Innovation Solution
A practical LMMSE (PLMMSE) channel estimation algorithm is developed, which uses LS estimation as a basis, calculates an N-dimensional channel autocorrelation matrix, and computes a weight matrix to improve channel estimation accuracy, reducing reliance on prior statistic information and achieving performance comparable to LMMSE.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LMMSE channel estimation is used to improve estimation accuracy, then channel estimation performance is improved, but calculation complexity increases and prior statistic information is required
Solution Approach 1:
The patent uses Least Square (LS) estimation as a simple, computationally inexpensive preliminary step to obtain an initial channel estimation result. This disposable preliminary estimation serves as input for the subsequent weight matrix calculation, allowing the system to achieve LMMSE-level performance without directly computing the complex LMMSE solution from scratch.
Solution Approach 2:
The patent divides the channel estimation process into two distinct stages: first performing LS estimation to obtain an initial result, then using this result to calculate a weight matrix for refinement. This segmentation allows the complex estimation problem to be broken down into manageable steps, reducing overall computational burden while maintaining accuracy.
2Measurement precision
If LMMSE channel estimation is used to improve estimation accuracy, then channel estimation performance is improved, but prior statistic information is required
Solution Approach 1:
The system performs self-service by calculating the weight matrix using the autocorrelation matrix derived from the received pilot signal itself, rather than requiring external prior statistic information. The preliminary LS estimation result is used to construct the weight matrix, making the system adaptive to current channel conditions without needing pre-stored statistical data.
Solution Approach 2:
The patent performs preliminary LS estimation before calculating the final channel estimation result. This preliminary action provides the necessary input for weight matrix calculation, enabling the system to adapt to current channel conditions in real-time without requiring prior statistic information about the channel.
3Device complexity
If DFT-Based channel estimation is used to reduce calculation complexity, then calculation complexity is reduced, but noise filtering and interference suppression performance deteriorates
Solution Approach 1:
The patent introduces feedback by using the autocorrelation matrix of the channel, which is derived from the received pilot signal and preliminary estimation results, to calculate the weight matrix. This feedback mechanism allows the system to adapt to actual channel conditions and optimize noise filtering performance based on real-time signal characteristics.
Solution Approach 2:
The patent changes the estimation parameter by introducing a weight matrix that is calculated based on the autocorrelation matrix and preliminary LS estimation results. This parameter change allows the system to dynamically adjust the estimation process to optimize performance for current channel conditions, improving noise filtering compared to fixed DFT-Based methods.
Data Source
AI summary
Disclosed are a method and device for estimating a channel in a multiple-receiving antenna system. The method comprises: a channel estimation is performed on a received pilot signal by using a least square channel estimation algorithm to obtain an estimation value HLS; an N′-dimensional channel autocorrelation matrix formula (I) and a channel frequency domain autocorrelation matrix Ri from a transmitting antenna i to any receiving antenna are acquired, and a weight matrix W which is descrambled and denoised is calculated, wherein i=1, 2, . . . , NT, and NT is the number of antennae at a transmitting end; the estimation value HLS is corrected by using the weight matrix W to obtain a corrected estimation value HP-LMMSE. The method and device of the disclosure may obtain the performance as close as possible to the LMMSE technology through less prior statistic information, and have a simple implementation manner.